Recent changes to par_grid

par_grid modified by A. Smith

A. Smith — Thu, 08 Sep 2016 10:54:46 -0000

--- v6
+++ v7
@@ -3,4 +3,4 @@

 3 entries for a 3D spectrum:

-    par.grid=\[10 10 5\];
+    par.grid=[10 10 5];

par_grid modified by A. Smith

A. Smith — Thu, 08 Sep 2016 10:54:35 -0000

--- v5
+++ v6
@@ -2,4 +2,5 @@
 The FitSpec function breaks the spectrum into sub-spectra, which are fitted individually, and then periodically recombined. The number of sub-spectra is determined by ‘par.grid’, which is a 1 X N vector, where N is the number of dimensions. Each element of ‘par.grid’ specifies how many parts that dimension should be separated into, so that the total number of sub-spectra is the product of the values in ‘par.grid’. Large values in ‘par.grid’ will accelerate fitting, since the size of each sub-spectrum is reduced, but peaks that overlap strongly into other sections will be fit less accurately. Usually, default settings are sufficient, but spectra with peaks that are quite broad in comparison to the spectrum resolution may not fit well with the default values of ‘par.grid’. The best remedy for this is to process with fewer data points, but if this is not possible due to a mix of broad and narrow peaks, then one may reduce the values of ‘par.grid’.

 3 entries for a 3D spectrum:
+
     par.grid=\[10 10 5\];

par_grid modified by A. Smith

A. Smith — Thu, 08 Sep 2016 10:54:14 -0000

--- v4
+++ v5
@@ -2,4 +2,4 @@
 The FitSpec function breaks the spectrum into sub-spectra, which are fitted individually, and then periodically recombined. The number of sub-spectra is determined by ‘par.grid’, which is a 1 X N vector, where N is the number of dimensions. Each element of ‘par.grid’ specifies how many parts that dimension should be separated into, so that the total number of sub-spectra is the product of the values in ‘par.grid’. Large values in ‘par.grid’ will accelerate fitting, since the size of each sub-spectrum is reduced, but peaks that overlap strongly into other sections will be fit less accurately. Usually, default settings are sufficient, but spectra with peaks that are quite broad in comparison to the spectrum resolution may not fit well with the default values of ‘par.grid’. The best remedy for this is to process with fewer data points, but if this is not possible due to a mix of broad and narrow peaks, then one may reduce the values of ‘par.grid’.

 3 entries for a 3D spectrum:
-par.grid=\[10 10 5\];
+    par.grid=\[10 10 5\];

par_grid modified by A. Smith

A. Smith — Thu, 08 Sep 2016 10:27:19 -0000

--- v3
+++ v4
@@ -2,4 +2,4 @@
 The FitSpec function breaks the spectrum into sub-spectra, which are fitted individually, and then periodically recombined. The number of sub-spectra is determined by ‘par.grid’, which is a 1 X N vector, where N is the number of dimensions. Each element of ‘par.grid’ specifies how many parts that dimension should be separated into, so that the total number of sub-spectra is the product of the values in ‘par.grid’. Large values in ‘par.grid’ will accelerate fitting, since the size of each sub-spectrum is reduced, but peaks that overlap strongly into other sections will be fit less accurately. Usually, default settings are sufficient, but spectra with peaks that are quite broad in comparison to the spectrum resolution may not fit well with the default values of ‘par.grid’. The best remedy for this is to process with fewer data points, but if this is not possible due to a mix of broad and narrow peaks, then one may reduce the values of ‘par.grid’.

 3 entries for a 3D spectrum:
-par.grid=[[10 10 5]];
+par.grid=\[10 10 5\];

par_grid modified by A. Smith

A. Smith — Thu, 08 Sep 2016 10:26:59 -0000

--- v2
+++ v3
@@ -2,4 +2,4 @@
 The FitSpec function breaks the spectrum into sub-spectra, which are fitted individually, and then periodically recombined. The number of sub-spectra is determined by ‘par.grid’, which is a 1 X N vector, where N is the number of dimensions. Each element of ‘par.grid’ specifies how many parts that dimension should be separated into, so that the total number of sub-spectra is the product of the values in ‘par.grid’. Large values in ‘par.grid’ will accelerate fitting, since the size of each sub-spectrum is reduced, but peaks that overlap strongly into other sections will be fit less accurately. Usually, default settings are sufficient, but spectra with peaks that are quite broad in comparison to the spectrum resolution may not fit well with the default values of ‘par.grid’. The best remedy for this is to process with fewer data points, but if this is not possible due to a mix of broad and narrow peaks, then one may reduce the values of ‘par.grid’.

 3 entries for a 3D spectrum:
-par.grid=[10 10 5];
+par.grid=[[10 10 5]];

par_grid modified by A. Smith

A. Smith — Thu, 08 Sep 2016 10:26:42 -0000

--- v1
+++ v2
@@ -1,3 +1,4 @@
+###par.grid
 The FitSpec function breaks the spectrum into sub-spectra, which are fitted individually, and then periodically recombined. The number of sub-spectra is determined by ‘par.grid’, which is a 1 X N vector, where N is the number of dimensions. Each element of ‘par.grid’ specifies how many parts that dimension should be separated into, so that the total number of sub-spectra is the product of the values in ‘par.grid’. Large values in ‘par.grid’ will accelerate fitting, since the size of each sub-spectrum is reduced, but peaks that overlap strongly into other sections will be fit less accurately. Usually, default settings are sufficient, but spectra with peaks that are quite broad in comparison to the spectrum resolution may not fit well with the default values of ‘par.grid’. The best remedy for this is to process with fewer data points, but if this is not possible due to a mix of broad and narrow peaks, then one may reduce the values of ‘par.grid’.

 3 entries for a 3D spectrum:

par_grid modified by A. Smith

A. Smith — Thu, 08 Sep 2016 10:26:24 -0000

The FitSpec function breaks the spectrum into sub-spectra, which are fitted individually, and then periodically recombined. The number of sub-spectra is determined by ‘par.grid’, which is a 1 X N vector, where N is the number of dimensions. Each element of ‘par.grid’ specifies how many parts that dimension should be separated into, so that the total number of sub-spectra is the product of the values in ‘par.grid’. Large values in ‘par.grid’ will accelerate fitting, since the size of each sub-spectrum is reduced, but peaks that overlap strongly into other sections will be fit less accurately. Usually, default settings are sufficient, but spectra with peaks that are quite broad in comparison to the spectrum resolution may not fit well with the default values of ‘par.grid’. The best remedy for this is to process with fewer data points, but if this is not possible due to a mix of broad and narrow peaks, then one may reduce the values of ‘par.grid’.

3 entries for a 3D spectrum:
par.grid=[10 10 5];